The vitamin A metabolite retinoic acid has long been recognized to induce a broad spectrum of biological effects. A variety of structural analogues of retinoic acid have been synthesized that also have been found to be bioactive. Some, such as Retin-A® (registered trademark of Johnson & Johnson) and Accutane® (registered trademark of Hoffmann-LaRoche), have found utility as therapeutic agents for the treatment of various pathological conditions. Metabolites of vitamin A and their synthetic analogues are collectively herein called “retinoids”.
Synthetic retinoids have been found to mimic many of the pharmacological actions of retinoic acid. However, the broad spectrum of pharmacological actions of retinoic acid is not reproduced in full by all bioactive synthetic retinoids.
Medical professionals have become very interested in the medicinal applications of retinoids. Among their uses approved by the FDA is the treatment of severe forms of acne and psoriasis. A large body of evidence also exists that these compounds can be used to arrest and, to an extent, reverse the effects of skin damage arising from prolonged exposure to the sun. Other evidence exists that these compounds may be useful in the treatments of a variety of severe cancers including melanoma, cervical cancer, some forms of leukemia, and basal and squamous cell carcinomas. Retinoids have also shown an ability to be efficacious in treating premalignant cell lesions, such as oral leukoplakia, and to prevent the occurrence of malignancy.
Use of the retinoids is associated with a number of significant side effects. The most serious of these is that, as a class, they are among the most potent teratogens known. Teratogens are compounds that cause severe birth defects during specific periods of fetal exposure. Other side effects include irritation of the tissues treated, which can be so severe that patients cannot tolerate treatment.
Various investigations have been undertaken to elucidate the structure-activity relationships governing the abilities of synthetic retinoids to induce the various pharmacological consequences of retinoic acid exposure. This has been a complicated task, however, since the assays available to investigators have been bioassays, carried out either in intact animals or in isolated tissues. Technical constraints have often dictated the use of different small animal species for different assays. Interpretation of results has been complicated by possible pharmacokinetic and metabolic effects and possible species differences in the receptors involved. Nevertheless, definite differences in the pharmacological effects of various synthetic retinoids have been observed.
Major insight into the molecular mechanism of retinoic acid signal transduction was gained in 1988. Prior to that time, several high abundance cellular retinoid binding proteins were incorrectly inferred to be the signal transducing receptors for retinoic acid. In 1988, a member of the steroid/thyroid hormone intracellular receptor superfamily (Evans, Science, 240:889-95 (1988)) was shown to transduce a retinoic acid signal (Giguere et al., Nature, 330:624-29 (1987); Petkovich et al., Nature, 330: 444-50 (1987)). This unexpected finding related retinoic acid to other non-peptide hormones and elucidated the mechanism of retinoic acid effects in altering cell function. It is now known that retinoids regulate the activity of two distinct intracellular receptor subfamilies; the Retinoic Acid Receptors (RARs) and the Retinoid X Receptors (RXRs).
The first retinoic acid receptor identified, designated RAR-alpha, acts to modulate transcription of specific target genes in a manner which is ligand-dependent, as has been shown to be the case for many of the members of the steroid/thyroid hormone intracellular receptor superfamily. The endogenous low-molecular-weight ligand upon which the transcription-modulating activity of RAR-alpha depends is all-trans-retinoic acid. Retinoic acid receptor-mediated changes in gene expression result in characteristic alterations in cellular phenotype, with consequences in many tissues manifesting the biological response to retinoic acid. Two additional genes closely related to RAR-alpha were recently identified and were designated RAR-beta and RAR-gamma and are very highly related (Brand et al., Nature, 332:850-53 (1988); Ishikawa et al., Mol. Endocrin., 4:837-44 (1990)). In the region of the retinoid receptors which can be shown to confer ligand binding, the primary amino acid sequences diverge by less than 15% among the three RAR subtypes or isoforms. All-trans-retinoic acid is a natural ligand for the retinoic acid receptors (RARs) and is capable of binding to these receptors with high affinity, resulting in the regulation of gene expression. The newly-discovered retinoid metabolite, 9-cis-retinoic acid, is also an activator of RARs.
A related but unexpected observation was made recently (Mangelsdorf et al., Nature, 345:224-29 (1990)), in which another member of the steroid/thyroid receptor superfamily was also shown to be responsive to retinoic acid. This new retinoid receptor subtype has been designated Retinoid X Receptor (RXR), because certain earlier data suggested that a derivative of all-trans-retinoic acid may be the endogenous ligand for RXR. Like the RARs, the RXRs are also known to have at least three subtypes or isoforms, namely RXR-alpha, RXR-beta, and RXR-gamma, with corresponding unique patterns of expression (Manglesdorf et al., Genes & Devel., 6:329-44 (1992)).
Although both the RARs and RXRs respond to all-trans-retinoic acid in vivo, the receptors differ in several important aspects. First, the RARs and RXRs are significantly divergent in primary structure (e.g., the ligand binding domains of RARα and RXRα have only 27% amino acid identity). These structural differences are reflected in the different relative degrees of responsiveness of RARs and RXRs to various vitamin A metabolites and synthetic retinoids. In addition, distinctly different patterns of tissue distribution are seen for RARs and RXRs. For example, in contrast to the RARs, which are not expressed at high levels in the visceral tissues, RXRα mRNA has been shown to be most abundant in the liver, kidney, lung, muscle and intestine. Finally, the RARs and RXRs have different target gene specificity. For example, response elements have recently been identified in the cellular retinal binding protein type II (CRBPII) and apolipoprotein AI genes which confer responsiveness to RXR, but not RAR. Furthermore, RAR has also been recently shown to repress RXR-mediated activation through the CRBPII RXR response element (Manglesdorf et al., Cell, 66:555-61 (1991)). These data indicate that two retinoic acid responsive pathways are not simply redundant, but instead manifest a complex interplay. Recently, Heyman et al. (Cell, 68:397-406 (1992)) and Levin et al. (Nature, 355:359-61 (1992)) independently demonstrated that 9-cis-retinoic acid is a natural endogenous ligand for the RXRs. 9-cis-retinoic acid was shown to bind and transactivate the RXRs, as well as the RARs, and therefore appears to act as a “bifunctional” ligand.
In view of the related, but clearly distinct, nature of these receptors, ligands which are more selective for the Retinoid X Receptor subfamily would be of great value for selectively controlling processes mediated by one or more of the RXR isoforms, and would provide the capacity for independent control of the physiologic processes mediated by the RXRs. Ligands which preferentially affect one or more but not all of the receptor isoforms also offer the possibility of increased therapeutic efficacy when used for medicinal applications.
The entire disclosures of the publications and references referred to above and hereafter in this specification are incorporated herein by reference.